 Hello, my name is Francisca Malfil and my name is Sophie Simons. We work at the Center for Medical Genetics Gantt, which is located at the campus of the Gantt University Hospital, Belgium. In this video, we wish to present our manuscript in which we studied defects in the carboxytermal propeptide domain, or in short, the C-propeptide domain of Tak-1 Procologen, and provide a molecular explanation on how these defects cause a brittle bone disease, osteogenesis imperfecta. Osteogenesis imperfecta, or OI, is characterized by a reliable degree of bone fragility, with susceptibility to bone fractures and limb deformities, growth deficiency, blue sclery, dentinogenesis imperfecta and hearing impairment. In approximately 85% of all OI patients, a mutation is identified in one of the genes, including Tak-1 Procologen, Tak-1-E-1, and Tak-1-E-2 gene. Tak-1 Procologen, the principal protein in bone and many other connective tissues, is synthesized as a precursor molecule, Procologen, which consists of 2-pro-alpha-1 and 1-pro-alpha-2 chain. Each pro-alpha chain typically contains a central triple helix domain, which is flanked by two globular extensions, the aminoterminal and carboxyterminal propeptides. The c-pro-peptides, which have a highly conserved sequence, are essential for correct selection and association of the 3-pro-alpha chains, thereby initiating triple helix formation. After completion of the triple helix, the Procologens are cleaved, thereby triggering the assembly of collagen molecules into fibros. As such, the c-pro-peptides have crucial roles in tissue growth and repair. Interestingly, only a small fraction of all known type 1 Procologen mutations reside in the c-pro-peptide domain. With our study, we aim to provide insights into the genotype-phenotype correlations underlying type 1 Procologen c-pro-peptide mutations. We reviewed the clinical, molecular and biochemical data on 30 c-pro-peptide variants identified at the Center for Medical Genetics Gantt. We compared our data with previously published data on c-pro-peptide mutations. The total number of type 1 Procologen c-pro-peptide variants identified is 83, which constitutes approximately 6.5% of all type 1 Procologen defects. Clear genotype-phenotype correlations emerge from this study. Pathogenic variants that do not allow association of the mutant chain generally result in mild OI. These can be divided into two groups. Firstly, pathogenic variants that generate a premature termination codon and that induce non-sosmediated mRNA decay lead to the production of less type 1 Procologen. In addition, a few pathogenic variants have been reported that lead to production of mutant Pro Alpha-1-1 chains with an altered c-pro-peptide sequence, but which are rapidly degraded inside itself and not included in the type 1 Procologen heterotrimer. A second group are pathogenic C-1A1 variants that result in the production of stable mutant Pro Alpha chains that can be incorporated into the type 1 Procologen heterotrimers, thereby delaying their folding. Intracellular accumulation of these misfolded chains has been shown to activate the unfolded protein response and increase endoplasmic reticulum stress. These defects generally cause the most severe OI phenotypes. The third group are pathogenic variants located in the c-pro-peptide of the Pro Alpha-2 chain. Generally, these are associated with a milder OI phenotype when compared to their counterparts in the Pro Alpha-1 chain. Recent elucidation of the crystal structure of the type 3 Procologen has shown that the c-pro-peptide domain, or c-p3 for short, has the shape of a flower with three petals, each corresponding to an individual polypeptide chain connected together at the base. Within the base region, each chain contains a tightly bound calcium ion, shown in blue, an interaction between chains or stabilized by interchained sulfate bonds, which are shown in yellow. Because the amino acid sequence of c-p3 is highly similar to that from Procologen type 1 or short c-p1, we use the structure of c-p3 to predict the expected consequences of OI-related miscense variants in c-p1. Two examples are shown. The first variant leads to the replacement of a polar trionin at position 1431 by a hydrophobic isoleucine. Trionin 1431 normally forms hydrogen bonds, which lies in 1433 and arginine 1436, which are involved in chain tremorization. Replacement of trionin by isoleucine will disrupt this hydrogen bonds and destabilize a trimer consistent with the OI type 4 phenotype. Not far away, the hydrophobic isoleucine 1439 is replaced by a polar trionin. Since isoleucine 1439 is buried in a strongly hydrophobic environment, replacement by a polar amino acid will also be destabilizing, consistent with the observed severe OI phenotype. As such, this study demonstrates that the crystal structure of c-p3 is a reliable tool to predict phenotypic severity for most c-p1A1 c-propeptite miscense variants. With this, we would like to thank our co-authors, all referring physicians and last but not least, all OI patients.